This project focuses on the study of membranes, proteins and carbohydrates by molecular dynamics computer simulation. Force field (FF) development continued for three classes of compounds: lipids, carbohydrates, and ethers. All involved extensive ab initio calculations on small model compounds, followed by extensive molecular dynamics simulations (MD) on the target systems to ensure that a variety of target data were reproduced. The CHARMM carbohydrate FF was extended to disaccharides (the monosaccaharide was completed last year). This enables simulations of numerous polysaccharides, though links to 5 membered rings and some functional groups still need to be incorporated. The 3D-IPS/DFFT method developed by Wu and Brooks (of LCP) was demonstrated to be highly accurate for liquid/liquid and liquid/vapor interfaces, and lipid bilayers and monolayers. Analysis of results from bilayers and monolayers indicated that the large bilayer surface tension arising from the C27r lipid FF is incorrect. This paved the way for a revised lipid FF with substantially lower surface tensions. The new lipid FF, C36, will be submitted for publication shortly, and is already in use by a number of research groups. A coarse-grained (CG) model for polyethylene oxide and polyethylene glycol was developed using simulations from the all-atom model developed last year. Simulations of 9, 18, 27, 36, 44, 67, 76, 90, 112, 135, and 158-mers of the CG PEO (442 <Mw <6998) at low concentration in water show the experimentally observed transition from ideal chain to real chain behavior near Mw = 2000, and agree with a number of other observables. The simulations also indicate that the concentration dependence of the apparent radius of gyration of PEG observed in scattering experiments primarily arises from interchain interference;i.e., the calculated radius of gyration at 21 and 148 mg/cc is equal. Simulations of PEO grafted to a nonadsorbing surface yield a mushroom to brush transition that is well described by the Alexander-de Gennes formalism. A consistent method for calculating the thickness of the brush and mushroom regimes was developed, and provides a physical interpretation of the Flory radius defined by deGennes. Applications of the CG to PEGylated lipid bilayers, and to self assembly of vesicles and bicelles is underway. Molecular dynamics simulations of sorbitol revealed a previously unappreciated internal hydrogen bond not possible in glucose. This internal bond explains the pattern of NMR T1 relaxation times, which differs qualitatively from that of linear alkanes, and may be related to unexplained deviations from ideality observed for polyols. MD simulations of phosphatidylinositol (4,5)-bisphosphate (PIP2) and phosphatidylinositol (3,4,5)-trisphosphate (PIP3) at low (physiological) concentration in bilayers indicate that the inositol rings are tilted 40 with respect to the bilayer surface. This tilt differs substantially from the experimentally observed value (near 90) of PIPs at high concentrations. Electrostatic potentials evaluated from Poisson-Boltzmann (PB) calculations show a strong dependence of potential height and ring orientation. These surfaces are well above the background height of for negatively charged cell membranes, as would be expected for lipids involved in cellular signaling. PB calculations on microscopically flat bilayers yield similar maxima as the MD-based (microscopically rough) systems, though show less fine structure. These results support the utility of a rigid/flat bilayer model for PB based studies of PIP2 and PIP3, as long as the orientations are judiciously chosen.